public inbox for libabigail@sourceware.org
 help / color / mirror / Atom feed
* [Bug default/29568] New: abidiff - misses diffs in highly connected ABI
@ 2022-09-13 10:36 gprocida at google dot com
  2022-09-13 11:05 ` [Bug default/29568] abidiff - performance regression gprocida at google dot com
  2022-09-13 11:20 ` gprocida at google dot com
  0 siblings, 2 replies; 3+ messages in thread
From: gprocida at google dot com @ 2022-09-13 10:36 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29568

            Bug ID: 29568
           Summary: abidiff - misses diffs in highly connected ABI
           Product: libabigail
           Version: unspecified
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: default
          Assignee: dodji at redhat dot com
          Reporter: gprocida at google dot com
                CC: libabigail at sourceware dot org
  Target Milestone: ---

Created attachment 14332
  --> https://sourceware.org/bugzilla/attachment.cgi?id=14332&action=edit
test case files

Hi Dodji.

While doing some performance tests, I found an abidiff issue (with an older
version of libabigail).

The attached test case includes two XML files which are almost but not quite
identical.

I've written up the reproducible test case as a script (though there is a
chance that type ids won't quite match match what sed expects depending on the
exact build of abidw). In any case you can examine and diff the XML files "by
hand".

abidiff should report a diff but does not. --harmless makes no difference.

I'm not sure if the "connectedness" of the ABIs is significant.

This "old" abidiff performance is fine here, BTW, but current libabigail is
very slow. So you could also use this example for performance testing abidiff.
I don't actually know how long it takes to run as it's still running as I type
this. I suppose it's possible that current libabigail does find the diff... in
which case this is a performance bug instead. I'll follow up if/when it
terminates.

Regards,
Giuliano.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [Bug default/29568] abidiff - performance regression
  2022-09-13 10:36 [Bug default/29568] New: abidiff - misses diffs in highly connected ABI gprocida at google dot com
@ 2022-09-13 11:05 ` gprocida at google dot com
  2022-09-13 11:20 ` gprocida at google dot com
  1 sibling, 0 replies; 3+ messages in thread
From: gprocida at google dot com @ 2022-09-13 11:05 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29568

gprocida at google dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|abidiff - misses diffs in   |abidiff - performance
                   |highly connected ABI        |regression

--- Comment #1 from gprocida at google dot com ---
The good news is that current abidiff finds the differences.

The bad news is that it takes a very long time to report them.

So I've retitled the bug.

There are no typedefs here.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 3+ messages in thread

* [Bug default/29568] abidiff - performance regression
  2022-09-13 10:36 [Bug default/29568] New: abidiff - misses diffs in highly connected ABI gprocida at google dot com
  2022-09-13 11:05 ` [Bug default/29568] abidiff - performance regression gprocida at google dot com
@ 2022-09-13 11:20 ` gprocida at google dot com
  1 sibling, 0 replies; 3+ messages in thread
From: gprocida at google dot com @ 2022-09-13 11:20 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29568

--- Comment #2 from gprocida at google dot com ---
$ time ~/android/libabigail/build/tools/abidiff --leaf-changes-only --harmless
demo/*xml
Leaf changes summary: 1 artifact changed
Changed leaf types summary: 1 leaf type changed
Removed/Changed/Added functions summary: 0 Removed, 0 Changed, 0 Added function
Removed/Changed/Added variables summary: 0 Removed, 0 Changed, 0 Added variable

'struct a199' changed:
  type size hasn't changed
  there are data member changes:
    type 'a198*' of 'a199::m000' changed:
      pointer type changed from: 'a198*' to: 'a199*'
    type 'a199*' of 'a199::m199' changed:
      pointer type changed from: 'a199*' to: 'a198*'

real    15m5.896s
user    14m56.284s
sys     0m9.557s

-- 
You are receiving this mail because:
You are on the CC list for the bug.

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2022-09-13 11:20 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-09-13 10:36 [Bug default/29568] New: abidiff - misses diffs in highly connected ABI gprocida at google dot com
2022-09-13 11:05 ` [Bug default/29568] abidiff - performance regression gprocida at google dot com
2022-09-13 11:20 ` gprocida at google dot com

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).